[Wien] Fe(111) surface - LAPW0 error

[Sun Yiyang]

Done. Thanks!

	-----Original Message----- 
	From: wien-admin at zeus.theochem.tuwien.ac.at on behalf of Torsten Andersen 
	Sent: Tue 9/7/2004 5:01 PM 
	To: wien at zeus.theochem.tuwien.ac.at 
	Cc: 
	Subject: Re: [Wien] Fe(111) surface - LAPW0 error
	
	

	Dear Yiyang,
	
	well, the "segmentation error" tells almost everything...
	
	what are your shell limits?
	in csh-like environments: limit
	in sh-like environments: ulimit -a
	
	you need to set memory and stack to unlimited. If it still does not
	help, you should probably add memory or swap space.
	
	Best regards,
	Torsten Andersen.
	
	Sun Yiyang wrote:
	> Dear Torsten,
	> 
	> Thank you for the diagnosis.
	> You're right. I missed the information in the .dayfile. It reports a "segmentation error" and that's all.
	> No error message is output on the screen. I changed the compiling option by using -O0 and
	> recompiled lapw0. But it doesn't help. I still got the same problem.
	> 
	> sgroup didn't complain anything. As for the .output0 file, I'm not sure it's ok or not.
	> (This is actually the first time I run WIEN2k on a system other than the tutorial. I'm trying to
	> make comparison with the results from a pseudopotential calculation which I'm more familiar with.)
	> I wonder if other people can't see my attachment either. I attach them again in txt format below.
	> 
	> --
	> Yiyang
	> 
	>       -----Original Message-----
	>       From: wien-admin at zeus.theochem.tuwien.ac.at on behalf of Torsten Andersen
	>       Sent: Tue 9/7/2004 2:27 PM
	>       To: wien at zeus.theochem.tuwien.ac.at
	>       Cc:
	>       Subject: Re: [Wien] Fe(111) surface - LAPW0 error
	>      
	>      
	>
	>       Dear Mr. Sun,
	>      
	>       you should probably use a smaller mixing factor (0.01-0.05), but since
	>       it stops already in the first lapw0 this can not be the problem (yet).
	>      
	>       Sorry, but I didn't see any attachment... but here are a few suggestions:
	>      
	>       Is there any information in case.dayfile?
	>       How much of the expected output in case.output0 is present?
	>       Any information on the screen when it terminates?
	>       Does it work if you compile with -O0?
	>       Does sgroup complain about something/anything?
	>      
	>       Best regards,
	>       Torsten Andersen.
	>      
	>       Sun Yiyang wrote:
	>       > Dear WIEN2k users,
	>       >
	>       > I'm trying to do a surface calculation on Fe(111) surface.
	>       > I followed the introduction by S.Cottenier to set up
	>       > the linearization energies (see attached .in1).
	>       > I used a 13-layer slab and a vacuum region equivalent
	>       > to 13 bulk interlayer spacings (see attached .struct).
	>       > The lattice constant I got is 5.3346 a.u.
	>       > Running init_lapw is ok, where I used -7.0 Ry for the
	>       > separation energy. Mixing factor is changed to 0.10.
	>       > I used 200 k-points (10x10x1). Calculation is spin-polarized.
	>       > But when I runsp_lapw. It stops at the first LAPW0
	>       > with an error line in LAPW0.error
	>       > "Error in LAPW0". No other error message.
	>       > Can anybody give some idea on this? Many thanks!
	>       >
	>       > Yiyang

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 8058 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20040908/27e31c54/attachment.bin